852 



SCIENCE 



[N. S. Vol. XL. No. 1041 



cloud was selected with the idea of noting its 

 changes of shape and something of its rate 

 of movement. The cloud selected advanced to 

 a point almost exactly overhead, then began to 

 melt away. In a space of less than five min- 

 utes it had entirely disappeared. Another 

 .and yet another did the same. Finally, an 

 unusually large cloud was selected; but this, 

 too, disappeared on reaching approximately 

 the same point. All advanced in orderly pro- 

 cession from the west till, overhead, they 

 reached a lane of clear sky, then melted away. 

 , This lane of clear sky, several miles wide, 

 stretched away northeastward to the horizon, 

 following very closely the course of the valley 

 of the Illinois Eiver, and southwestward over 

 the river valley for 4 or 5 miles, after which it 

 gave place to the usual cumulus clouds. To 

 the east of the valley the cumulus clouds ap- 

 peared once more and continued to the hori- 

 zon. These relations were observed to persist 

 throughout the greater part of the afternoon. 



A possible explanation of the phenomenon 

 which suggested itself at the time is here re- 

 corded as a working hypothesis to be consid- 

 ered in connection with similar occurrences 

 which may from time to time be described. To 

 make this explanation clear, a brief descrip- 

 tion of the geography of the region is neces- 

 sary. 



The valley of the Illinois Eiver here is a 

 flat-bottomed trough from 4 to 10 miles wide, 

 bordered by relatively sharp bluffs, and is 

 sunk some 200 feet below the general upland 

 level of this part of the state. The upland is 

 mainly cultivated farmland, much of it at 

 this time of year bare after the wheat, hay and 

 oats harvest. The river bottom, on the other 

 hand, east of the point of observation and to 

 the northeast as far as could be seen, is largely 

 either swampy, with several lakes, or forest- 

 covered. Four or five miles to the southwest, 

 however, in the neighborhood of Beardstown, 

 a considerable portion of the river bottom has 

 been reclaimed and is given over to agricul- 

 ture. 



The explanation suggested is that over the 

 upland farms numerous convection currents 

 gave rise to cumulus clouds, while over the 



swamp and forest lands of the river bottom 

 convection currents were subordinate or, per- 

 haps, absent; that consequently, this cooler 

 belt over the bottomlands not only failed to 

 produce new cumulus clouds but also tended to 

 become the channel of descent for some of the 

 air which had been rising by convection from 

 the surrounding hotter lands. On reaching 

 such a belt of descending air, the clouds should 

 be expected to melt away as they were drawn 

 downward and to leave a zone of clear sky 

 over the area of descending air. The width of 

 the valley — 4 to 10 miles — as compared with 

 the height of the clouds — probably about one 

 mile — should give ample opportunity for differ- 

 ences in heating to become effective in modi- 

 fying the air currents and therefore the be- 

 havior of the clouds. 



The presence of the cumulus clouds over the 

 reclaimed parts of the bottomlands near 

 Beardstown is thought to be significant in con- 

 nection with the above explanation, since these 

 would doubtless be heated nearly as effectively 

 as the upland. 



Other possible explanations of the phe- 

 nomenon might be suggested, but it seems idle 

 to speculate further until more observations 

 of a similar nature have been made. 



John L. Eioh 



University of Illinois 



cyanide op potassium in trees 

 To THE Editor op Science : I note an inter- 

 esting contribution to Science in the issue of 

 October 9, on the subject of cyanide of potas- 

 sium taken up by trees when put into holes in 

 the same. I wish to report that this chemical 

 is the chief basis of treatment by a firm in 

 AUentown, Pa., doing an extensive business 

 in some of the Eastern States, claiming to 

 render trees immune from attacks by all in- 

 sects and diseases, and also to fertilize them. 

 The process was originated by a man named 

 Kleckner, and is now continued by a company 

 called the Fertilizing Scale Company, of AUen- 

 town, Pa. Their theory is that a tree can be 

 given medicine, as well as food, by placing the 

 same in capsules and fastening these in inci- 

 sions under the bark. While the chief insect 



